Preliminary RF rankings

1. Babe Ruth
2. Hank Aaron
3. Mel Ott
4. Frank Robinson
5. Roberto Clemente
6. Pete Rose
7. Shoeless Joe Jackson
8. Reggie Jackson - literally one spot behind Shoeless Joe in my all-positions rankings
9. Paul Waner
10. Harry Heilmann
11. Larry Walker
12. Sam Crawford
13. Tony Gwynn
14. Reggie Smith
15. Dwight Evans
16. Sammy Sosa
17. Bobby Bonds
18. Elmer Flick
19. Bobby Abreu
20. Gary Sheffield
21. Vladimir Guerrero
22. Dave Winfield
23. Willard Brown
24. Enos Slaughter
25. Willie Keeler
26. Pete Browning (I have as a CF?)
27. Sam Thompson

I think the list is missing Al Kaline. It looks like we had King Kelly here last time - I have him here to, but I think listing him as a catcher is justified.

Prelim:
1. Babe Ruth
2. Henry Aaron
3. Mel Ott
4. Frank Robinson
5. Al Kaline
6. Roberto Clemente
7. Pete Rose
8. Reggie Jackson
9. Larry Walker
10. Paul Waner
11. Joe Jackson
12. Enos Slaughter
13. Dwight Evans
14. Harry Heilmann
15. Gary Sheffield
16. Elmer Flick
17. Tony Gwynn
18. Sam Crawford
19. Reggie Smith
20. Vladimir Guerrero
21. Bobby Bonds
-- Kiki Cuyler
22. Willie Keeler
23. Sammy Sosa
-- Tommy Henrich
-- Brian Giles
24. Bobby Abreu
25. Dave Winfield
-- Harry Hooper
26. Willard Brown
27. Pete Browning
-- A bunch of guys
28. Sam Thompson

A few thoughts - Sam Thompson is one of the weakest HoM choices. Probably has a fair case as the single weakest choice. Not a fan of Brown or Browning either, but those are both guys who probably deserve more attention on my end.

I'm definitely lower on Crawford or Winfield than most people. They are reasonably similar to Manny Ramirez, who I am also low on. I'm going to guess my Enos Slaughter ranking is higher tham most voters - not sure why my system likes him so much, but it's something to look into.

Just to add to Jaack's notes:

I don't think there should be any way that Pete Browning is in the RF group. He was in CF last time, and according to BB-Ref, he only played 35 games in RF, as opposed to 477 in LF and 490 in CF.

And I would probably put King Kelly in with the RF group as well. He played 742 games in RF as opposed to 584 at C. And he played catcher more towards the end of his career, when the seasons were longer than at his debut, so he definitely has an even greater season fractions at RF than C compared to just games played.

Kcgard2, I don't see Al Kaline, accidentally omission, and if so, where would he place? And is Enos Slaughter at #24 with WWII credit

Jaack, I need to do a full work up and on mobile, but I have Sheffield ahead, maybe Slaughter is somewhere around #13. He deserves prime/peak WWII credit.

Agree with M Brinkley, Browning in CF and Kelly in RF makes sense.

Also, Willard Brown should be in CF as well. According to Seamheads, he only played 7 games in RF, but 310 in CF (138 in LF).

BB-Ref has it 107-253-23 games going from left field to right.

I mostly lurk and rarely comment, but I have to say that I don't see how Ott ranks above Frank Robinson. They are only 3.6 WAR apart on career, and Robinson played in a much more competitive environment. Even as a Giants fan, I would reverse the rankings I see above.

I mostly lurk and rarely comment, but I have to say that I don't see how Ott ranks above Frank Robinson. They are only 3.6 WAR apart on career, and Robinson played in a much more competitive environment. Even as a Giants fan, I would reverse the rankings I see above.

There's a pretty substantial gap for me bewteen Ott and Robinson. I think Ott accrued a bit more value than Robinson (I see the difference at more around 8-9 WAR) and his value is a bit more focused - basically all of that gap is in Ott's top 6 seasons. And while Ott did play in a segregated league, he also played under a shortened schedule - I do account for both, but it kind of ends up as a wash.

Robinson ended up with about 400 extra PAs, and I guess I don't see giving Ott another 400 as sufficient to overcome the segregation problem. But I admit that if you do that it's pretty much a wash.

I added Kaline and Kelly and removed Browning from the list. This matches how we had people assigned last time.

Willard Brown was also ranked with the CF in the previous ranking as well.

Will update. If Willard Brown is in CF that makes seven CF who are not PHoM, plus Cool Papa Bell who wouldn't be if I had to do it all over again. We've elected 31 center fielders.

I'm working on re-doing my rankings, but at cursory glance I would have 6 HoM CF that are not in my PHoM as well (WB wouldn't be one of them, however, but Cool Papa Bell would be), although I would have 3 PHoM, not HoM CF that would partially replace them.

Removing Willard Brown and Pete Browning, adding Al Kaline and King Kelly:

1. Babe Ruth
2. Hank Aaron
3. Mel Ott - Ott above Robinson because for the all-time rankings I'm not doing any timelining (a pennant is a pennant). Adjusting competition levels would alter my rankings considerably.
4. Frank Robinson
5. Roberto Clemente
6. Al Kaline
7. Pete Rose
8. Shoeless Joe Jackson
9. Reggie Jackson - literally one spot behind Shoeless Joe in my all-positions rankings
10. Paul Waner
11. Harry Heilmann
12. Larry Walker
13. Sam Crawford
14. Tony Gwynn
15. Reggie Smith
16. Dwight Evans
17. Sammy Sosa
18. Bobby Bonds
19. Elmer Flick
20. Bobby Abreu
21. Gary Sheffield
22. Vladimir Guerrero
23. Dave Winfield
24. Enos Slaughter - this placement includes war credit
25. King Kelly
26. Willie Keeler
27. Sam Thompson - the only RF who is not pHOM

8. Mefisto Posted: January 26, 2023 at 10:44 PM (#6114725)
I mostly lurk and rarely comment, but I have to say that I don't see how Ott ranks above Frank Robinson. They are only 3.6 WAR apart on career, and Robinson played in a much more competitive environment. Even as a Giants fan, I would reverse the rankings I see above.

A few things to help Ott:

Defense rated relatively better to Robinson in other metrics, sometimes significantly.
Worse home/road splits for F Rob
SIGNIFICANTLY worse relative clutch figures for Frank.

Prelim:
1. Babe Ruth

2. Henry Aaron

3. Mel Ott

4. Frank Robinson
5. Al Kaline - from previous post, Robinson comes down far enough and Kaline is strong enough by others that he almost nips Frank.

6. Reggie Jackson
7. Roberto Clemente

8. Larry Walker
9. Joe Jackson
10. Pete Rose

11. Paul Waner
12. Gary Sheffield
13. Dave Winfield
14. Sam Crawford
-Ichiro Suzuki-
15. Enos Slaughter
16. Dwight Evans

17. Harry Heilmann
- Mookie Betts -
18. Reggie Smith

19. King Kelly
20. Vladimir Guerrero
21. Elmer Flick
22. Tony Gwynn

In:
23. Bobby Abreu
24. Bobby Bonds

Borderline In:
-Brian Giles-
-Jack Clark-
25. Willie Keeler
-Bryce Harper-

Borderline Out:
-Harry Hooper-
-Kiki Cuyler-
-Tommy Henrich-
26. Sammy Sosa

Next Up:
Gavvy Cravath
Sam Rice

Quick estimate of group:
Darryl Strawberry
Rocky Colavito
Dave Parker
Giancarlo Stanton
Nelson Cruz
Tony Oliva
Jose Bautista
Jose Canseco
28. Sam Thompson

Preliminary Rankings
1. Babe Ruth. Clear #1, any way I look at it.
2. Henry Aaron. Equally clear #2.
3. Frank Robinson. In context, a little better than Mel Ott.
4. Mel Ott. Didn’t have a lot of competition for top power hitter in the National League in the 1930s.
5. Roberto Clemente. Best defensive right fielder in the HoM.
6. Al Kaline. Quietly great all-around.
7. Pete Rose. Sometimes I forget how good at baseball Pete Rose actually was.
8. Reggie Jackson. Top power hitter during a tough period for power hitters.
9. Larry Walker. Would be up with Clemente and Kaline with fewer injuries.
10. Joe Jackson. Would be up with Clemente and Kaline if his career had continued.
11. Sam Crawford. Tough pick between Crawford and Heilmann. I give Crawford the nod for excelling during the deadball era, where Heilmann needed the live-ball boost to excel: in context, Crawford stands out a bit more.
12. Harry Heilmann.
--Ichiro Suzuki (with Japan credit only roughly estimated)
13. Paul Waner.
14. Tony Gwynn.
15. Gary Sheffield. Offensively, a good comp for Reggie Jackson, but defensively, not so much. From Sheffield here to Winfield at 26, there's much distance between the players.
16. Bobby Abreu. Being very good at everything is great.
17. Enos Slaughter. Hard to place. This is with war credit. His contextual ranking is quite high, but his raw score is rather low. This is a compromise spot.
18. King Kelly. Contemporary reputation was higher than the reality, but he was still a great player in his prime. With a stronger finish, he’d be in the Heilmann/Waner/Gwynn tier.
19. Willie Keeler. Higher than I expected.
20. Reggie Smith. Smith and Evans are a tricky pair in my system. Very similar in raw value, and a lot overlap in their careers, but Evans goes into the 1980s and Smith into the 1970s. Evans ranks higher overall among 1980s players than Smith among 1970s players, but Smith ranks significantly higher among 1970s position players than Evans does among 1980s position players, so he gets the nod for now.
21. Dwight Evans.
22. Elmer Flick. One of the few hitters to be highly successful in the depths of the deadball era.
23. Sammy Sosa. Lots of arguments against him, but unless you fully buy into the clutch hitting stats, still a solid, lower-tier HoMer.
24. Bobby Bonds. A solid, lower-tier HoMer. Strong in many aspects of the game.
--Brian Giles
25. Vladimir Guerrero. Others seem to have him a bit higher. Not sure what the difference is. It may be that the differences lie more in the assessments of other players than in assessments of Guerrero.
26. Dave Winfield. I grew up seeing him as a superstar, but he wasn’t quite that. Still, he’s above my in-out line.
--Hurley McNair, Heavy Johnson, Jose Cruz, Harry Hooper, and Kiki Cuyler are around in here, probably.
27. Sam Thompson. Only elected rightfielder below my in-out line. He’s not a bad choice, in that I don’t think there are any better players from the 1890s that should have been elected instead of Thompson, but I think we went a little deeper into the 1890s than we should have.

1. Babe Ruth
2. Hank Aaron
3. Mel Ott
4. Frank Robinson
5. Al Kaline
6. Roberto Clemente
7. Reggie Jackson
8. Larry Walker
9. Paul Waner
10. Sam Crawford
11. Reggie Smith
12. Gary Sheffield
13. Dwight Evans
14. Shoeless Joe Jackson
15. Bobby Bonds
16. Vladimir Guerrero
17. Sammy Sosa
18. Bobby Abreu
19. Dave Winfield
20. Elmer Flick
21. Tony Gwynn
22. Enos Slaughter
23. Harry Heilmann
24. King Kelly

This is my PHoM cutoff line

25. Willie Keeler
26. Sam Thompson

My PHoM includes Heavy Johnson and Gavvy Cravath - they both fall between Gwynn and Slaughter.

25. Vladimir Guerrero. Others seem to have him a bit higher. Not sure what the difference is. It may be that the differences lie more in the assessments of other players than in assessments of Guerrero.

Vlad, Abreu, and Sosa are all extremely similar to me - offensive is about the same in quality, and all of them have a lot of room for interpretation on their defense. I rate Vlad the best, but mostly because I'm not hedging there at all. He has a better peak than Abreu and none of Sosa's weirdness with contextual stats - he's delightfully middle of the road for a bottom quartile HoMer to a point where I'm fairly confident he isn't a mistake in a way I'm not with Sosa or Abreu.

"Extremely similar" is definitely what I see here. The bottom half of the right field set flattens out more than the left fielders did. I had a very hard time settling on an order here, and my rankings may still shuffle quite a bit. I doubt I'm gping to change anything in left field, though, before I send my ballot.

cookie, you seem to be missing Pete Rose.

Prelim ballot

1) Ruth
2) Aaron
3) Ott
4) Frank Robinson - nothing controversial so far
5) Paul Waner
6) Al Kaline
7) Sam Crawford
8) Tony Gwynn - surprisingly good defender
9) Pete Rose
10) Reggie Jackson
11) Harry Heilmann
12) Larry Walker
13) Enos Slaughter - 2 seasons war credit
14) Gary Sheffield
15) Roberto Clemente - only 70 WAR and 33 WAPA puts him here. Big contrast with Sheffield on how they created value.
16) Willie Keeler - excellent defensive numbers
17) Elmer Flick
18) King Kelly
19) Joe Jackson - I delete any value he had after he cheated in the World Series applying the lifetime ban to the moment after the rule was broken
20) Reggie Smith
21) Dwight Evans
22) Dave Winfield
23) Sam Thompson - I have him at 31 WAPA, need to dig deeper on this one
-- Brian Giles
24) Bobby Abreu
-- Gavy Cravath
25) Sammy Sosa
26) Vlad Guerrero
27) Bobby Bonds - not PHoM but not too far below the line. He's right next to Kiki Cuyler and Chuck Klein on my list

DL from MN, could you explain a bit more about your placement of Clemente? You are assigning Clemente only 70 WAR, which seems a bit low considering that Baseball Reference has him at 94.8 and Fangraphs at 80. Given that you are also listing WAPA, I am wondering if that's Dan R WAR? If so, that's pretty seriously underrating Clemente, I think. From some basic measures of offensive and defensive value, it seems like Clemente should fit into the group that you have in the 5-8 range in your rankings, alongside the other 3000-hit, doubles-triples-power rightfielders:

Waner, 1926-45, 134 OPS+ in 10766 PA, 23 rField, 4 rBaser+rDP
Kaline, 1953-74: 134 OPS+ in 11597 PA, 154 rField, 27 rBaser+rDP
Clemente, 1955-72: 130 OPS+ in 10212 PA, 205 rField, 33 rBaser+rDP
Gwynn, 1982-2001: 132 OPS+ in 10232 PA, 6 rField, 29 rBaser+rDP

How is it that Clemente does not align with this group, but instead ranks with Gary Sheffield and Willie Keeler? Especially given that he and Kaline are exact contemporaries, and Kaline played in the weaker league, what factors would lead him to have much higher value than Clemente, given their similar offensive and defensive performance?

1. Babe Ruth
2. Hank Aaron
3. Mel Ott
4. Frank Robinson
5. Roberto Clemente
6. Al Kaline
7. Joe Jackson
8. Harry Heilmann
9. Larry Walker
10. Sam Crawford
11. Paul Waner
12. Tony Gwynn
13. Reggie Jackson
14. Pete Rose
15. Enos Slaughter
16. Gary Sheffield
17. Elmer Flick
18. Reggie Smith
19. Dwight Evans
20. King Kelly
21. Bobby Bonds
22. Willie Keeler
23. Bobby Abreu
24. Sammy Sosa
25. Vladimir Guerrero
26. Dave Winfield
27. Sam Thompson

@ Chris Cobb In #24, hopefully Kiko can join us soon, but his Win Loss Records are much lower on Clemente as well: https://baseball.tomthress.com/StatTables/PlayerStats.php?id=clemr101&w0=0.333333333333&w5=0&wl=0.333333333333&ws=0.333333333333&p0=0.333333333333&p1=0.333333333333&p2=0.333333333333

This placement helps me move Kaline and Reggie ahead, but I'm hard pressed to drop him below 7th.

His rankings help elevate Winfield into the mid-tier as well. He was an excellent road relative hitter that gives him a boost too.

Dan R on Clemente:

BWAA2 BRWAA2 FWAA2 Rep2 WARP2 WAPA2
-1.8 -0.1 0.3 -0.7 -1 -2.7
0.7 -0.1 -0.1 -0.7 1.2 -0.8
-1.8 -0.2 1 -0.6 -0.4 -2.1
0 0.2 1.8 -0.7 2.7 0.7
-0.2 0 0.3 -0.5 0.5 -1.1
2.1 -0.1 0.2 -0.6 2.9 0.6
4.3 0.1 0.5 -0.6 5.6 3.2
1.3 0 1.2 -0.6 3.1 1
3.1 0.4 -0.3 -0.7 3.8 1.6
4.6 0.1 0.8 -0.7 6.3 3.7
3.5 0.3 1.5 -0.7 6 3.6
4.6 0.1 1.4 -0.8 6.9 4.3
6.1 0.3 0.4 -0.7 7.5 5.1
3.7 0 1.8 -0.6 6.1 4
5.1 0.2 0 -0.7 5.9 3.8
3.6 0.1 0.4 -0.5 4.7 3
3.3 0 0.6 -0.6 4.4 2.5
2.1 0.5 0.9 -0.5 4 2.4
TOTALS (ignoring seasons 1-3)
47.2 2.2 11.5 -9.5 70.4 38.4

I noticed I should delete Clemente's first 3 seasons because he is well below average. His 5th season reduces his value in my system but I include it because season 4 has value enough to cover it.

Deleting seasons 1-3 moves him just ahead of Heilmann

cookie, you seem to be missing Pete Rose.

I had him as a LFer. In RF, he's between Jackson and Bonds.

@16: Thanks.

long time voter, but also long time no vote, chiming in with my surprise at consensus of Ott over Frank Robinson.

Yes, the stats per year slightly favor Ott. But timeline and integraiton are important, aren't they? And *surely* a discount is warranted for the war years when Ott was still racking up good numbers. Yes, we should also adjust a wee bit for season length.

Post-season stats don't move the needle much. MVP and HoF voting, both clearly in Frank's favor.

Best seasons? WAR has Ott ahead; Win Shares has Robinson with the top 2 figures.

The WAR numbers ding Robinson more for position adjustment. I don't know if that is justified.

All in all, I plainly disagree.

And *surely* a discount is warranted for the war years when Ott was still racking up good numbers.

I have trouble with how to deal with war years. I see some voters who don't want to give war credit and also want to give a discount to people who played. That will systematically undervalue a generation of players.

I've come to the conclusion that I don't have any good handle on quality of play so I wouldn't know how to systematically adjust for it beyond the standard deviations in run scoring.

Was Frank Robinson the better baseball player? Probably. Did Ott provide more value to the teams he played for at the time? Probably. I feel like I should be answering the second question for Hall of Merit voting.

Best seasons? WAR has Ott ahead; Win Shares has Robinson with the top 2 figures.

Some quick hitters for Ott vs Robinson.

Ott:
B-R: 110.8
B-G - 100.2
T-T - ~141
Clutch: -4
WPA: 80.5
Home/Road: 106/94
Playoffs: 901 OPS in 69 PA.

Robinson:
B-R: 107.2
B-G - 105.2
T-T - ~105
Clutch: -15.6
WPA: 73.0
Home/Road: 110/91
Playoffs: 887 OPS in 149 PA.

Was Frank Robinson the better baseball player? Probably. Did Ott provide more value to the teams he played for at the time? Probably. I feel like I should be answering the second question for Hall of Merit voting.

My sense would be that accurately answering the second question is necessary but not sufficient for answering the first question, which is the more fundamental one. We consider factors in addition to "how much value did a player provide to the teams he played for at the time" in a lot of different ways--giving credit for time missed due to military service or labor actions being two of the more obvious examples. The members of the electorate do not all share the same views of the most proper ways to consider factors that go beyond "how much value did a player provide to the teams he played for at the time?" but I'd guess we all do it in some fashion. Adjusting for the competitive context when comparing the values created in two different contexts is one of those ways, just as giving war credit is.

I am not trying in this post to argue that one should or shouldn't give war credit or make adjustments for competitive context, but I am trying to point out that whatever we are doing, we are generally not just answering the question of who provided more value to the teams he played on at the time.

I have trouble with how to deal with war years. I see some voters who don't want to give war credit and also want to give a discount to people who played. That will systematically undervalue a generation of players.

I agree that the approach described above is problematic, but I would have take just as much issue with giving war credit and not discounting the remaining players. One alternative artificially limits the total amount of value awarded, the other artificially inflates it.

The fact that Robinson has a bigger home/road split than Ott is highly counterintuitive, because Ott is often described as a player who benefited overwhelmingly from his home park due to having hit 323 of his 511 HR in the Polo Grounds. But he also hit for a higher average (.311-.297) on the road, and hit vastly more doubles and triples away from home (306-181, 51-21). It's an excellent example of how park effects are narratively oversimplified to be home runs only, ignoring (for instance) the large foul territory in the Polo Grounds.

After adding postseason data:

Rose moves up 1
Slaughter moves up 2
Dwight Evans moves ahead of Reggie Smith

I would have take just as much issue with giving war credit and not discounting the remaining players. One alternative artificially limits the total amount of value awarded, the other artificially inflates it.

Well, giving war credit and discounting players are operations of differing complexity. It's one thing to estimate what one player would likely have done if he hadn't been removed from play by military service, but to establish the amount of value reduction for the whole league based on the absence of all the players who were in military service is a much more difficult operation. (It's also psychologically more pleasant to restore value to players than to devalue what they actually accomplished.) It's also a much smaller degree of change for much more calculative effort.

So I am sympathetic to those who give war credit but who decide that calculating penalties isn't worth it.

When you're talking about discounting the list is actually pretty small. Musial, Appling, Keller, Boudreau, Gordon, Vaughan in 1943 but that should be a small discount. Musial, Newhouser, Boudreau, Bob Johnson, Bucky Walters, Vern Stephens, Bobby Doerr in 1944. Newhouser, Hack and Vern Stephens in 1945. The Negro Leaguers (Gibson, Campanella, Leonard, Brown, Paige) don't get dinged for playing. Newhouser gets affected the most because it's his peak, although his best seasons are 45-46 where discounting should be small. Instead of calculating a discount the best option would probably be regressing those seasons based on the individual player's performance in surrounding seasons.

That's a helpful list of players, DL, and using regression as an adjustment strategy is a good idea, and not one I had considered before.

With respect to the Negro-League players, I think that in principle discounting would be just as appropriate there: the Negro Leagues were affected by the draft just as the National and American Leagues were. However, discerning anything about an appropriate level of discount is heavily complicated by the presence of other factors, such as the large number of Black players who played in Mexico in the early 1940s. The question of how competition levels in the Negro Leagues were affected by having to compete with the MxL for players, followed by depletion of the talent pool by the draft has ramifications that go beyond the assessment of the players in the NeL during the 1941-45 period because a large portion of the data that we have from players who played in the Negro Leagues and then later in the National and American Leagues comes from that period. If competition levels were reduced during that time by these factors, then the application of any competition adjustment derived from comparing player performance in the 1941-45 NeL to their performance in the 1947- NL and AL to other periods of Black Baseball becomes problematic without further adjustments to factor out the impact of the draft. I think this is the sort of thing Dr. Chaleeko may be working through now. It's an immensely complicated problem.

Aren't the NGL MLEs already regressed using surrounding seasons? I don't think he directly translates year-by-year data.

I can't speak with authority about Dr. Chaleeko's current methodology. I believe he uses regression, but there is also a competition adjustment in the translation to a National League context. Because 1945 NeL would be translated into 1945 NL, it may be that the "wartime discount" would be the same for both contexts and thus would not need to be examined within the process of creating MLEs. If a wartime discount for NL/AL players is handled through regression, then presumably the regression already incorporated into the MLE methodology for NeL players would take care of it, and one would have no need to apply a wartime discount to NeL MLEs. However, in both cases, if some or all of the seasons providing the baseline for regression are part of the set of seasons that need discounting, it may not be quite as simple as that--I don't know. I do believe that the question of the impact of wartime conditions on competition levels does become pertinent for the establishment of the overall competition adjustments for the NeL to NL/AL, because in that case the comparisons are being made between play in wartime seasons and later play in non-wartime seasons. The competition study that Dr. Chaleeko is working on may not approach the assessment of competition levels using that kind of comparisons, so it may not be relevant to his project.

Those comparisons were/are part of the way I established competition adjustments for my own MLEs, so the issue of the depression of competition levels in the NeL during WW2 is certainly pertinent to my own approach, for whatever that approach is worth.

just looking to confirm:

these were our 20 choices in (real time) 2008:
Hank Aaron
Roberto Clemente
Sam Crawford
Dwight Evans
Elmer Flick
Tony Gwynn
Harry Heilmann
Joe Jackson
Reggie Jackson
Al Kaline
Willie Keeler
King Kelly
Mel Ott
Frank Robinson
Pete Rose
Babe Ruth
Enos Slaughter
Sam Thompson
Paul Waner
Dave Winfield

so I believe this time we have added:
Bobby Abreu
Bobby Bonds
Vladimir Guerrero
Gary Sheffield
Reggie Smith
Sammy Sosa
Larry Walker

I believe that the list in #42 is the correct one - updated prelim taking that in mind

1. Babe Ruth
2. Henry Aaron
3. Mel Ott
4. Frank Robinson
5. Al Kaline
6. Roberto Clemente
7. Pete Rose
8. Reggie Jackson
9. Larry Walker
10. Paul Waner
11. Joe Jackson
12. Enos Slaughter
13. Dwight Evans
14. Harry Heilmann
15. Gary Sheffield
-- Ichiro Suzuki (This is w/ rough NPL credit)
16. Elmer Flick
17. Tony Gwynn
18. Sam Crawford
19. Reggie Smith
-- Mookie Betts
20. King Kelly
21. Vladimir Guerrero
22. Bobby Bonds
-- Kiki Cuyler
23. Willie Keeler
24. Sammy Sosa
-- Tommy Henrich
-- Brian Giles
25. Bobby Abreu
26. Dave Winfield
-- Harry Hooper
-- Bryce Harper
-- Gavy Cravath
-- Dave Parker
-- Rocky Colavito
-- Sam Rice
-- Jack Clark
-- Darryl Strawberry
-- Tony Oliva
27. Sam Thompson

I don't feel confident enough in my Hurley McNair or Heavy Johnson rankings right now to put them in here, but I feel quite confident that they are better than Thompson.

Here is my preliminary ranking, which shouldn't be overly controversial.

I enjoyed the Ott vs. Robinson discussion above and have flipped them in my ranking several times since I started working on this. Ultimately, I'm Team Robinson (at least for the moment) for two reasons:

- I think it's reasonable to, when in doubt, favor the integration-era player.

- I definitely believe in regressing performance during years when many top players were at war. As someone above pointed out, if you're going to credit guys who weren't there (as I think you should), you almost have to debit those who were. It was simply a weaker league. I have Newhouser as a borderline candidate and Doerr and Bob Johnson on the outside in part for this reason. That discount obviously doesn't knock Ott too far down, but at the margins here it's a factor.


1. Babe Ruth
2. Henry Aaron
3. Frank Robinson
4. Mel Ott
5. Roberto Clemente
6. Al Kaline
7. Pete Rose
8. Reggie Jackson
9. Joe Jackson
10. Paul Waner
11. Harry Heilmann
12. Tony Gwynn
13. Larry Walker
14. Sam Crawford
15. Enos Slaughter
16. Dave Winfield
17. Dwight Evans
18. Reggie Smith
19. Vladimir Guerrero
20. Gary Sheffield
21. Sammy Sosa
22. King Kelly
23. Elmer Flick
24. Willie Keeler
25. Bobby Abreu
26. Bobby Bonds
27. Sam Thompson

I didn't expand my thoughts above, but you expressed my view very well.

I see people were having trouble posting to the ballot thread again. Usually posting in any open thread will unlock a closed thread for me. Also, put a prelim in here and it will get counted if you can't post later. Third, you can send me a personal email and I'll add the ballot.

I don't see a RF ballot thread, even in the HOM section of the site (where LF discussion and ballot are posted next to each other).

I haven't opened the RF ballot yet. Will do in about 2 weeks.

Interesting to see Chris and I way out of consensus on Keeler. Keeler was not a controversial pick at all when he was first selected. Dan R's data covers his entire career and he gives 63.6 WARP2 and 37.4 WAPA2. He also rates as one of the best defensive RF in the HoM just behind Clemente.

It looks like early players are doing less well in this round of rankings than in the last one: the two biggest drops in the left field cohort were Jesse Burkett, Jimmy Sheckard, and Charlie Jones. The evaluation of Keeler among the right fielders may be parallel to the evaluation of the early leftfielders.

It is also seems likely that there is going to be less agreement among the rankings for rightfielders than was the case for leftfielders. In leftfield, the rankings fell into several clear cohorts. That seems to be less true for right field, where almost half of the players are very closely grouped together, with small enough differences in their career values that even modest divergences among evaluation systems will produce quite different rank orderings, whether those divergences be the way peak is calculated or which comprehensive metrics are used or how different historical periods are integrated.

Putting this set of players into a rank order is going to exaggerate the magnitude of the differences between them, so evaluations of players may look more divergent, one ballot to another, than they actually are.

I am very curious to see how this ranking will shake out.

As far as Keeler goes - he is clearly behind Delahanty/Clarke/Burkett tier for COF in that era. The next tier for me is Flick/Sheckard/Keeler/Kelley. It's hard for me to get Keeler>Sheckard. The bats are quite similar in terms of value, but Sheckard's glove looks to be a step ahead of Keeler. I like Flick better than both, but can see the argument either way there. That makes Keeler the sixth best OF of his era - impressive considering how strong the era is, but at the same time it's hard for me to rate him too highly.

I have him next to Sosa, which feels right enough - they both place similarly in the OF hierarchy when you account for league size, and have oddly similar profiles for guys 100 years apart (good bat in a good hitting, OF rich era. One really big year with a uniquely impressive trad stat achievement. Good, not great glove).

I'm not sure if RF is just less hierarchical than LF, or perhaps if the middle class is just more controversial. Gwynn. Crawford, Sheffield, and Slaughter all seem like guys with a lot of room for disagreement, whereas with Goose Goslin or Billy Williams aren't.

Ranking by MMP score (15 points for unanimous first place for a season)

Right Fielder Points
Babe Ruth 204.89
Henry Aaron 146.40
Frank Robinson 73.79
Mel Ott 67.59
Joe Jackson 48.01
(Mookie Betts 47.09)
Elmer Flick 42.58
Al Kaline 33.43
Reggie Jackson 31.79
(Albert Belle 31.29)
Harry Heilmann 29.24
Paul Waner 26.98
Tony Gwynn 25.68
Dwight Evans 25.67
(Dave Parker 24.71)
Sammy Sosa 23.91
Sam Crawford 23.39
Gary Sheffield 22.10
Pete Rose 21.58
(Bryce Harper 20.86)
Roberto Clemente 20.59
Willie Keeler 16.59
Larry Walker 15.18
(Gavy Cravath 13.25)
(Brian Giles 12.35)
Dave Winfield 11.99
Enos Slaughter 11.75
(Kiki Cuyler 11.00)
Bobby Bonds 9.66
(Ichiro Suzuki 8.32)
Sam Thompson 7.65
Vladimir Guerrero 7.06
Reggie Smith 5.15
Bobby Abreu 1.77
King Kelly N/A

 51. Jaack Posted: February 07, 2023 at 02:46 PM (#6116045)
As far as Keeler goes - he is clearly behind Delahanty/Clarke/Burkett tier for COF in that era. The next tier for me is Flick/Sheckard/Keeler/Kelley. It's hard for me to get Keeler>Sheckard. The bats are quite similar in terms of value, but Sheckard's glove looks to be a step ahead of Keeler. I like Flick better than both, but can see the argument either way there. That makes Keeler the sixth best OF of his era - impressive considering how strong the era is, but at the same time it's hard for me to rate him too highly.

I have him next to Sosa, which feels right enough - they both place similarly in the OF hierarchy when you account for league size, and have oddly similar profiles for guys 100 years apart (good bat in a good hitting, OF rich era. One really big year with a uniquely impressive trad stat achievement. Good, not great glove).

I'm not sure if RF is just less hierarchical than LF, or perhaps if the middle class is just more controversial. Gwynn. Crawford, Sheffield, and Slaughter all seem like guys with a lot of room for disagreement, whereas with Goose Goslin or Billy Williams aren't.

I think I'm in agreement to pretty much EVERYTHING that you are saying here.

I'd add that I don't see where I can get Willie Keeler > King Kelly.

I could be convinced Vlad, Abreu, and Bonds are below Keeler, the just in/just out group here is significant.

I don't know what to do with Tony Gwynn.
He's excellent at Baseball-Reference, in clutch/situation hitting, and WPA.
He's borderline with Baseball Gauge.
He's short with Baseball Prospectus (though I question the use of DRC+, seems predictive but not quantitative).
He's woeful with Kiko's Stat.

Some mix of these and I have him in, but not as overwhelmingly as others (though looks like cookiedabookie is with me here too).

He's borderline with Baseball Gauge.

Just based on what I've looked at on Baseball Gauge WAR, it seems to put a sharply lower value on high-BA, low-power hitters than bWAR does (Ichiro, Wade Boggs and Rod Carew also take large hits in offensive value, for instance). I'm not familiar enough with the offensive methodology to understand why that is, but it makes me a bit skeptical of gWAR for that specific player type.

To Eric #54, Kiko can chime in, but his linear weights studies are in n agreement that power is underrated versus contact, so Boggs, Carew, and Gwynn end up lower compared with BR.

For what it's worth - this is probably not a flawless comparison because I think RE24 will factor in some things that aren't included in RBat, but here is the comparison between RE24 (changes in team run expectancy) and RBat (per bWAR, so the more favorable version for them) for those three guys:

Gwynn: 403 RBat; 538 RE24
Boggs: 441 RBat; 464 RE24
Carew: 407 RBat; 520 RE24

Gwynn and Carew are both pretty solid baserunners, but not 100-runs-above-average solid. Gwynn is 51st all-time in RE24 (Carew is 53rd); they're in the same neighborhood as George Brett and Edgar Martinez on that list. Gwynn goes even higher in WPA, moving up to 30th (55.96), right between Manny Ramirez and Mike Schmidt. (Carew's WPA ranking is about the same as his RE24; Boggs is notably lower than the others in both categories, especially WPA.) These aren't perfect stats, but they provide estimates of in-context run/win production in a way that RBat would not, and they seem to work in Gwynn's favor (and Carew's to a smaller but still notable extent).

I would very much be open to considering research to the contrary. For what it's worth, the link on Kiko's site leading to a comparison of player won-loss records to WAR appears to be a dead link. If a version of that info (or anything else relevant) is available I'll happily take a look.

I'd add that I don't see where I can get Willie Keeler > King Kelly.

Kelly is difficult to score. Too early for the system I'm using, catcher credit, short seasons. I think you may be right with Kelly > Keeler but it isn't by a whole lot.

I would very much be open to considering research to the contrary. For what it's worth, the link on Kiko's site leading to a comparison of player won-loss records to WAR appears to be a dead link. If a version of that info (or anything else relevant) is available I'll happily take a look.

If the link is dead, some of the old have been transferred to "Old Articles" format:
https://baseball.tomthress.com/OldArticles/UltimateStat.php

I would consider these as essential as well:
https://baseball.tomthress.com/OldArticles/Fielding.php

https://baseball.tomthress.com/OldArticles/eWORL_v_WAR.php

https://baseball.tomthress.com/OldArticles/HittingPositions_v_FieldingPositions.php

May take a minute for some of these to load...happy reading!

Given what's been posted about the highly varied assessments of Tony Gwynn, what I am led to wonder is why WPA and player won-lost records, which at least with respect to offense seem to be attempting to measure the same thing, have such different views of Gwynn's production? Kiko's argument is that linear weights, by focusing only on run values, undervalues the additional win value that the scoring certainty of the home run provides. That assertion provides some explanation of why Gwynn performs so poorly in his system. Does WPA, which purports to be presenting win values based on play-by-play context, not similarly recognize this value? I don't know the specifics of what's inside either calculation, but the divergence strikes me as odd.

I might say that Sammy Sosa provides an interesting inverse case. As is well known, his showing on WPA, RE24, and clutch measures is, well, very poor. It seems to be the case (I have not done a thorough study) that he looks worse by those measures than Gwynn does by player won-lost records. I would expect that player won-lost records would be similarly harsh with Sosa, but that isn't the case. His pWORL is 8 wins lower than his eWORL, but he still comes in around 50, nearly ten wins of pWORL better than Gwynn, and he goes up to around 58 on eWORL, nearly 15 ahead of Gwynn. If WPA and RE24 find that calculations of value derived from play-by-play data show Sosa's home runs to be empty, why do the player won-lost records, also based on play-by-play, reach such a different conclusion?

A player that might be interesting to compare to Gwynn is Bobby Abreu. Abreu had more power than Gwynn, but he wasn't a power hitter in the mold of Sammy Sosa or even Vlad Guerrero or Larry Walker. He has similar career length to Gwynn, similar total bases + walks (a little higher, but in a higher offense era), a little lesser by rate according to OPS+ and rbat+, shows well in WPA but not as good as Gwynn (trails him by 7 wins), but in player won-lost records, he has 38.3 eBatting Wins to Gwynn's 31.

I don't know why these large discrepancies in evaluations are showing up, and I'd like to know more.

I think for Sosa, player win-loss records see pretty much the same thing as WPA/RE24 - by eWORL, the bat is of HoM quality, but in a contextual setting, the bat looks relatively empty.

My one thought - both RE24 and WPA stop looking at a player once he gets on base, while I'm not sure how player win-loss records takes that into account. This is just a guess, but if Gwynn was coming around to score after his bases-empty single less often than expected by RE24/WPA, and player win-loss records is picking up on that, that would explain the discrepency. I'm not sure if that is what player win-loss records is picking up on, but it's a line of thought that makes some sense to me.

And for Gwynn, it's odd to say, but it's all about the offense.

Any defensive system I review, Gwynn is moderately above or below average for career, average mostly and spiking a couple of excellent years in his youth, then below average with a couple of poor years late career.

As I was looking at defense, I found Baseball Prospectus has rebranded as RDA, including statcast data from 2016 to present, looks like it retained the FRAA elements for earlier seasons.

I think for Sosa, player win-loss records see pretty much the same thing as WPA/RE24 - by eWORL, the bat is of HoM quality, but in a contextual setting, the bat looks relatively empty.

Looking more closely, I see that you are right, at least with respect to WPA, which can be compared to batting wins derived from bWAR pretty easily. The difference for Sosa between pWORL and eWORL is 8.5 wins. The difference between Sosa's WPA (25.0) and his rbat converted into WAA (30.6) is 5.6 wins, which is similar in magnitude to the pWORL eWORL difference (and the larger difference in player won-lost records might be accounted for by baserunning and fielding contextual differences not included in WPA).

But player won-lost records agree on Sosa, why do they differ on Gwynn so radically? Player won-lost records doesn't see all that much difference between context-Gwynn and context-neutral-Gwynn. His pWORL is 41.5, just a bit below his eWORL of 43.8. Player won-lost records eWins also sees context-neutral-Gwynn less favorably than does bWAR. Gwynn has 31.0 batting wins above average in eWins. His rbat of 403, however, converted to WAA would be about 41.1, so there is a 10-win difference there. WPA, however, goes way up from rbat's context-neutral evaluation of Gwynn's hitting. WPA gives him 56.0 wins above average, so it sees contextual evaluation raising Gwynn's value, rather than lowering it as in player won-lost records, and it appears to be raising it up from a higher baseline. I can understand more readily how the two systems using play-by-play data would have different context-neutral baselines for Gwynn than I can how contextual evaluation in one case could find Gwynn a somewhat less valuable hitter when context is applied while contextual evaluation in another case finds him a much more valuable hitter.

Again long time lurker, never poster here.

Right field was been a position where virtually all the candidates had decent to great peaks, long primes and in most cases excellent careers. I'm pretty sure that the consensus was that 19 of the 20 were pretty solidly in most personal halls of merit. Sam Thompson was the only outlier. Well since that time some 13-14 years ago, there have been 7 additions, they are primarily in the middle to the back of the pack as they are either from the backlog, or have been recently been elevated with the rise in WAR (baseball reference and fangraphs style). Of the 7, I think it's likely that only Larry Walker finishes in the top half, not to say that players like Sheffield were not front-loggers, he happened to come onto some stacked ballots, but as a whole the Right fielders are deep and have fewer holes in peak, prime and career than other positions.

Regarding positioning I'd expect Clemente and Kaline to bump up and could be 5-6, with Waner falling a bit and the balance of the 6 newbies other than walker falling in the bottom half (15-27) likely somewhere behind Heilmann (an all bat candidate). Rose and Joe Jackson were polarizing last time and I could see them being polarizing here, however again with peak, prime and career being considered, they are likely upper half (depending on potential penalties for Rose's hanging on, and potential demerits to Jackson's 1919-20 seasons).

Without too much further review... I'd rank them like this (career voter)
1. Ruth
2. Aaron
3. Ott
4. Robinson
5. Kaline
6. Clemente
7. Crawford
8. Waner
9. R. Jackson
10. Rose
11. J. Jackson
12. Walker
13. Kelly
14. Gwynn
15. Heilmann
16. Sheffield
17. Winfield
18. Flick
19. Slaughter
20. Evans
21. Guerrero
22. Keeler
23. Sosa
24. Smith
25. Abreu
26. Bonds
27. Thompson

Sorry further revision. Flick should be down 3 between Guerrero and Keeler. He's in but not by a huge amount.

Ballot thread is up

Welcome taxandbeerguy.

11. Michael J. Binkley's anxiety closet Posted: February 17, 2023 at 10:09 PM (#6117721)
Okay - I wanted to post this in the discussion thread first, but the site won't let me. I'm also posting later because I've been revising/simplifying my system. I'm still peak-heavy, but I've reduced some of the emphasis on rate, relying primarily on WAA/WAG with bonus points for MMMP/All-Star awards in my system plus a cWPA-based post-season bonus, if applicable.

1. Babe Ruth - my #1 player all-time
2. Henry Aaron
3. Mel Ott
4. Frank Robinson
5. Roberto Clemente
6. Reggie Jackson
7. Al Kaline
8. Joe Jackson
(Mookie Betts)
9. King Kelly
10. Pete Rose
11. Larry Walker
(Ichiro!)
12. Paul Waner
13. Tony Gwynn
14. Gary Sheffield
15. Elmer Flick
16. Dwight Evans
17. Harry Heilmann
18. Sam Crawford
19. Reggie Smith
20. Sammy Sosa
(Aaron Judge)
(Gavvy Cravath)
21. Vladimir Guerrero
22. Dave Winfield
(Heavy Johnson)
23. Willie Keeler
(Charlie Smith)
24. Sam Thompson
25. Bobby Bonds
(Brian Giles)
26. Bobby Abreu
27. Enos Slaughter

Good call out on Aaron Judge, he's risen to the borderline out area for me and I had forgot to note that.

With the new method, can you share a look under the hood, using King Kelly and Enos Slaughter.

King Kelly at #9 goes against consensus, and Enos Slaughter at #27 feels low, particulary when you bump for WWII credit.

Thanks for bringing up Slaughter. Apparently I forgot to give him war credit. He'd move up to 11th for me, but I'm not sure if I need to post a third ballot on the ballot thread

Yes, please post a corrected ballot to the ballot thread (ballot counters will understand).

With the new method, can you share a look under the hood, using King Kelly and Enos Slaughter.

King Kelly at #9 goes against consensus, and Enos Slaughter at #27 feels low, particulary when you bump for WWII credit.

Basic system review: I don't care about any seasons that a player had that were below average. Yes, those seasons may have value for his team, but they aren't what makes a player great. So I mostly only use WAA and WAG (I do use WAR, however, in my MMP voting/All-Star determinations that I use for bonus points in my system. I don't just take WAA as listed since WAR in BB-Ref is league-agnostic, but WAA is league-specific (e.g. in the UA a player's WAR was essentially the same as his WAG because the average player in the UA was replacement level for all of baseball). So I calculate a player's mWAR through averaging a number of systems, and then reverse engineer league-agnostic mWAA and mWAG (essentially All-Star level - 3 wins over average for a full season). I then add the player's mWAA and mWAG together, so WAG gets double counted because it's already included in mWAA, but that's because I want an emphasis on true greatness, not just compilers.

King Kelly - I have him with 40.8 mWAA and 9.7 mWAG. He also got bonus points for being the best position player in baseball in both 1879 and 1886, as well as being a 5-time All-Star (which for me is just a first-team best player at his position for that year, adjusting the number of pitchers for historical average number in a rotation for that year and including one relief pitcher for every year post WWII).

Enos Slaughter - I have him with, including war credit, 31.0 mWAA and only 3.1 mWAG. His only year for me above 6 mWAR was 1942 (7.7 mWAR, 5.3 mWAA, 2.1 mWAG) and he only had 2 other years above 5 mWAR (1948 and 1949). He's also only a 1-time All-Star for me.

It's very hard for a player to due really well in my system without at least one or two MVP level seasons. The player who does relatively best for me with less than 5 career mWAG appears to be another RF, Sam Crawford, whom I have at 18th among RF, which is obviously below consensus, but he's still a clear HoMer for me. Crawford only has 4.4 mWAG, but he racks up 39.4 mWAA because, although he doesn't have a single season above 7 mWAR for me, he has nine seasons between 5.1 and 6.9 mWAR.

The commenting glitch won't let me post in the ballot thread, so can this count as my official ballot please?

1. Babe Ruth
2. Hank Aaron
3. Mel Ott
4. Frank Robinson
5. Roberto Clemente
6. Al Kaline
7. Pete Rose
8. Sam Crawford
9. Reggie Jackson
10. Larry Walker
11. Tony Gwynn
12. Dwight Evans
13. Harry Heilmann
14. Enos Slaughter
15. Dave Winfield
16. Paul Waner
17. Reggie Smith
18. Joe Jackson
19. Sammy Sosa
20. Bobby Bonds
21. Gary Sheffield
22. Vladimir Guerrero
23. Bobby Abreu
24. Willie Keeler
25. Elmer Flick
26. King Kelly
27. Sam Thompson

All 27 of these RFs are in my PHOM, so I cast no aspersions on the names at the bottom of the list.

Ichiro, if eligible, would rank between Kaline and Rose (with credit for his pre-MLB play).